About
About Colourlab Ai

About Colourlab Ai

We've spent the last eight years asking a question that wasn't really being asked: what would it mean for software to understand an image the way a human perceives it? Not the mathematics of pixels, but the perception of light, color, and emotion that turns a frame into a feeling. Color Intelligence is the research company we built to answer it — and Colourlab Ai is what that answer looks like, so far.

We started in 2018. The question pulled us into color science, perceptual psychology, neural networks, and GPU architecture, and it produced things we didn't fully expect when we set out — our own log encoding (JPLog2), our own perceptual color space (Q Color Space), film emulation that responds to image content rather than baking it in, AI that learns a colorist's intent rather than copying their reference. Eight years of research doesn't sound long until you're inside it. From where we sit now, it's the foundation of everything in Colourlab Ai 4.

Before this work belonged to anyone else, it belonged to professionals. The world's leading studios, broadcasters, and post houses have been using Colourlab on theatrical features, episodic television, and high-end commercials for years now. That wasn't a marketing strategy — it was how we made sure the science actually worked. If a film colorist working on a Dolby Vision deliverable can trust our tools, the perceptual model underneath is real. That's the proof we needed before we were willing to put this technology into more hands.

Colourlab Ai 4 is the version where that work stops belonging only to specialists. The same perceptual engine that grades studio features now runs on a desktop fast enough to feel invisible — and, for the first time anywhere, on a phone in real time, with full AI color grading running entirely on-device. We recently announced Colourlab Ai Gen4 on the Honor V6, the first device to put professional color intelligence inside a smartphone with no cloud round-trip. That isn't a demo of where we're headed. It's a signal that the gap between Hollywood-grade and creator-grade is closing — and we intend to close it.

A few things we believe, that shape every decision in the software:

  • That perception matters more than mathematics. The mathematics has to be right — and ours is — but no one watching a film is doing arithmetic. They're feeling. The tools should be built around the felt experience, not the numerical one.
  • That AI belongs on your machine, not in someone else's data center. Creative flow doesn't survive a network round-trip. Intelligence has to be instant, private, and yours. Everything in Colourlab Ai 4 runs locally for that reason.
  • That AI should amplify a colorist, not replace one. We're not interested in software that grades for you. We're interested in software that removes the friction between your taste and the image — so the time you spend grading is spent on the parts only you can do.
  • That research takes the time it takes. Eight years of work doesn't compress into a sprint. We've chosen, repeatedly, to wait until the science was right rather than ship the trend that was loud at the time.

Where this goes next is in three directions at once. Color intelligence belongs on every device that makes images — phones, cameras, on-set monitors, editing platforms. It belongs inside the conversation, where a creator can describe a look and watch it appear. And it belongs in the parts of image-making that are still being invented — live production, streaming, real-time workflows that didn't exist when we started.

Colourlab Ai 4 is one step in a longer arc. We've spent eight years getting the foundations right. The next eight are about what gets built on top of them.

The work behind Colourlab is done by a small group of working colorists, scientists, and engineers. You can meet them here.

In This Section

Colourlab 4.0.11 BThe Team